Main

In your final repo, there should be an R markdown file that organizes all computational steps for evaluating your proposed Facial Expression Recognition framework.

This file is currently a template for running evaluation experiments. You should update it according to your codes but following precisely the same structure.

if(!require("EBImage")){
  install.packages("BiocManager")
  BiocManager::install("EBImage")
}

package 㤼㸱EBImage㤼㸲 was built under R version 4.0.3

if(!require("R.matlab")){
  install.packages("R.matlab")
}

package 㤼㸱R.matlab㤼㸲 was built under R version 4.0.4

if(!require("readxl")){
  install.packages("readxl")
}

package 㤼㸱readxl㤼㸲 was built under R version 4.0.3

if(!require("dplyr")){
  install.packages("dplyr")
}

package 㤼㸱dplyr㤼㸲 was built under R version 4.0.3

if(!require("readxl")){
  install.packages("readxl")
}

if(!require("ggplot2")){
  install.packages("ggplot2")
}

package 㤼㸱ggplot2㤼㸲 was built under R version 4.0.3

if(!require("caret")){
  install.packages("caret")
}

package 㤼㸱caret㤼㸲 was built under R version 4.0.4

if(!require("glmnet")){
  install.packages("glmnet")
}

package 㤼㸱glmnet㤼㸲 was built under R version 4.0.3

if(!require("WeightedROC")){
  install.packages("WeightedROC")
}

package 㤼㸱WeightedROC㤼㸲 was built under R version 4.0.4

if(!require("keras")){
  install.packages("keras")
}

package 㤼㸱keras㤼㸲 was built under R version 4.0.4

if(!require("tensorflow")){
  install.packages("tensorflow")
}

package 㤼㸱tensorflow㤼㸲 was built under R version 4.0.4

if(!require("magick")){
  install.packages("magick")
}

package 㤼㸱magick㤼㸲 was built under R version 4.0.4

library(R.matlab)
library(readxl)
library(dplyr)
library(EBImage)
library(ggplot2)
library(caret)
library(glmnet)
library(WeightedROC)
library(tensorflow)
library(keras)
library(magick)

Step 0 set work directories

set.seed(2020)
# setwd("~/Project3-FacialEmotionRecognition/doc")
# here replace it with your own path or manually set it in RStudio to where this rmd file is located. 
# use relative path for reproducibility

Provide directories for training images. Training images and Training fiducial points will be in different subfolders.

train_dir <- "../data/train_set/" # This will be modified for different data sets.
train_image_dir <- paste(train_dir, "images/", sep="")
train_pt_dir <- paste(train_dir,  "points/", sep="")
train_label_path <- paste(train_dir, "label.csv", sep="") 

Step 1: set up controls for evaluation experiments.

In this chunk, we have a set of controls for the evaluation experiments.

• (T/F) cross-validation on the training set
• (T/F) reweighting the samples for training set
• (number) K, the number of CV folds
• (T/F) process features for training set
• (T/F) run evaluation on an independent test set
• (T/F) process features for test set

run.cv <- TRUE # run cross-validation on the training set
sample.reweight <-  TRUE # run sample reweighting in model training
K <- 5  # number of CV folds
run.feature.train <- TRUE # process features for training set
run.test <- TRUE # run evaluation on an independent test set
run.feature.test <- TRUE # process features for test set

Using cross-validation or independent test set evaluation, we compare the performance of models with different specifications. In this Starter Code, we tune parameter lambda (the amount of shrinkage) for logistic regression with LASSO penalty.

lmbd = c(1e-3, 5e-3, 1e-2, 5e-2, 1e-1)
model_labels = paste("LASSO Penalty with lambda =", lmbd)

Step 2: import data and train-test split

#train-test split
info <- read.csv(train_label_path)
n <- nrow(info)
n_train <- round(n*(4/5), 0)
train_idx <- sample(info$Index, n_train, replace = F)
test_idx <- setdiff(info$Index, train_idx)

If you choose to extract features from images, such as using Gabor filter, R memory will exhaust all images are read together. The solution is to repeat reading a smaller batch(e.g 100) and process them.

n_files <- length(list.files(train_image_dir))

image_list <- list()
for(i in 1:100){
   image_list[[i]] <- readImage(paste0(train_image_dir, sprintf("%04d", i), ".jpg"))
}

Fiducial points are stored in matlab format. In this step, we read them and store them in a list.

#function to read fiducial points
#input: index
#output: matrix of fiducial points corresponding to the index
readMat.matrix <- function(index){
     return(round(readMat(paste0(train_pt_dir, sprintf("%04d", index), ".mat"))[[1]],0))
}

#load fiducial points
fiducial_pt_list <- lapply(1:n_files, readMat.matrix)
save(fiducial_pt_list, file="../output/fiducial_pt_list.RData")

Step 3: construct features and responses

• The follow plots show how pairwise distance between fiducial points can work as feature for facial emotion recognition.

  • In the first column, 78 fiducials points of each emotion are marked in order.
  • In the second column distributions of vertical distance between right pupil(1) and right brow peak(21) are shown in histograms. For example, the distance of an angry face tends to be shorter than that of a surprised face.
  • The third column is the distributions of vertical distances between right mouth corner(50) and the midpoint of the upper lip(52). For example, the distance of an happy face tends to be shorter than that of a sad face.

Figure1

feature.R should be the wrapper for all your feature engineering functions and options. The function feature( ) should have options that correspond to different scenarios for your project and produces an R object that contains features and responses that are required by all the models you are going to evaluate later.

• feature.R
• Input: list of images or fiducial point
• Output: an RData file that contains extracted features and corresponding responses

source("../lib/feature.R")
tm_feature_train <- NA
if(run.feature.train){
  tm_feature_train <- system.time(dat_train <- feature(fiducial_pt_list, train_idx))
  save(dat_train, file="../output/feature_train.RData")
}else{
  load(file="../output/feature_train.RData")
}

tm_feature_test <- NA
if(run.feature.test){
  tm_feature_test <- system.time(dat_test <- feature(fiducial_pt_list, test_idx))
  save(dat_test, file="../output/feature_test.RData")
}else{
  load(file="../output/feature_test.RData")
}

create a modified image list

img <- readImage(paste0(train_image_dir, sprintf("%04d", 1), ".jpg"))
print(img)

Image 
  colorMode    : Color 
  storage.mode : double 
  dim          : 1000 750 3 
  frames.total : 3 
  frames.render: 1 

imageData(object)[1:5,1:6,1]
          [,1]      [,2]      [,3]      [,4]      [,5]
[1,] 0.7098039 0.7176471 0.7215686 0.7176471 0.7098039
[2,] 0.6980392 0.7058824 0.7137255 0.7176471 0.7098039
[3,] 0.6901961 0.6980392 0.7098039 0.7137255 0.7137255
[4,] 0.6941176 0.7019608 0.7098039 0.7137255 0.7098039
[5,] 0.7019608 0.7098039 0.7137255 0.7137255 0.7058824
          [,6]
[1,] 0.7019608
[2,] 0.7058824
[3,] 0.7098039
[4,] 0.7058824
[5,] 0.6980392

setwd(train_image_dir)

The working directory was changed to C:/Users/evesu/Documents/GitHub/Spring2021-Project3-group-8/data/train_set/images inside a notebook chunk. The working directory will be reset when the chunk is finished running. Use the knitr root.dir option in the setup chunk to change the working directory for notebook chunks.

img.face <- sample(dir())

imgs <- list(NULL)

for(i in 1:length(img.face)){ 
  imgs[[i]]<- readImage(img.face[i])
 imgs[[i]]<- resize(imgs[[i]], 100, 100)
} 

create training tensors for kerasim

allimagesmod <- data.frame(index, imgs, values)

source("../lib/feature.R")

tm_feature_train <- NA

if(run.feature.train){
  tm_feature_train <- system.time(dat_train <- feature(fiducial_pt_list, train_idx))
  save(dat_train, file="../output/feature_train.RData")
}else{
  load(file="../output/feature_train.RData")
}

tm_feature_test <- NA

if(run.feature.test){
  tm_feature_test <- system.time(dat_test <- feature(fiducial_pt_list, test_idx))
  save(dat_test, file="../output/feature_test.RData")
}else{
  load(file="../output/feature_test.RData")
}

CNN

Step 4: Train a classification model with training features and responses

Call the train model and test model from library.

train.R and test.R should be wrappers for all your model training steps and your classification/prediction steps.

• train.R
  • Input: a data frame containing features and labels and a parameter list.
  • Output:a trained model
• test.R
  • Input: the fitted classification model using training data and processed features from testing images
  • Input: an R object that contains a trained classifier.
  • Output: training model specification
• In this Starter Code, we use logistic regression with LASSO penalty to do classification.

source("../lib/train.R") 
source("../lib/test.R")

For CNN’s Averagin is better? https://datascience.stackexchange.com/questions/47797/using-cross-validation-technique-for-a-cnn-model#:~:text=Any%20time%20you%20have%20models,tendency%20toward%20overfitting%20not%20underfitting.

feature_train = array(unlist(dat_train[, -6007]), dim = c(2400,152,3))
label_train = array(as.numeric(dat_train$label)) - 1 

• Train the model with the entire training set using the selected model (model parameter) via cross-validation.

# training weights

tm_train <- system.time(fit_train <- train(feature_train, label_train))

Model: "sequential_2"
________________________________________________________________________________________________
Layer (type)                               Output Shape                          Param #        
================================================================================================
conv1d_8 (Conv1D)                          (None, 150, 32)                       320            
________________________________________________________________________________________________
max_pooling1d_5 (MaxPooling1D)             (None, 75, 32)                        0              
________________________________________________________________________________________________
conv1d_7 (Conv1D)                          (None, 73, 64)                        6208           
________________________________________________________________________________________________
max_pooling1d_4 (MaxPooling1D)             (None, 36, 64)                        0              
________________________________________________________________________________________________
conv1d_6 (Conv1D)                          (None, 34, 64)                        12352          
________________________________________________________________________________________________
flatten_2 (Flatten)                        (None, 2176)                          0              
________________________________________________________________________________________________
dense_5 (Dense)                            (None, 64)                            139328         
________________________________________________________________________________________________
dense_4 (Dense)                            (None, 1)                             65             
================================================================================================
Total params: 158,273
Trainable params: 158,273
Non-trainable params: 0
________________________________________________________________________________________________
Epoch 1/10
53/53 - 0s - loss: 4.1443 - accuracy: 0.7458 - auc: 0.5064
53/53 - 1s - loss: 4.1443 - accuracy: 0.7458 - auc: 0.5064 - val_loss: 0.5203 - val_accuracy: 0.7944 - val_auc: 0.5781
Epoch 2/10
53/53 - 0s - loss: 0.5338 - accuracy: 0.7994 - auc: 0.5414
53/53 - 1s - loss: 0.5338 - accuracy: 0.7994 - auc: 0.5414 - val_loss: 0.5154 - val_accuracy: 0.7958 - val_auc: 0.6327
Epoch 3/10
53/53 - 0s - loss: 0.5001 - accuracy: 0.8030 - auc: 0.5759
53/53 - 1s - loss: 0.5001 - accuracy: 0.8030 - auc: 0.5759 - val_loss: 0.4910 - val_accuracy: 0.8014 - val_auc: 0.6590
Epoch 4/10
53/53 - 0s - loss: 0.4742 - accuracy: 0.8077 - auc: 0.6384
53/53 - 1s - loss: 0.4742 - accuracy: 0.8077 - auc: 0.6384 - val_loss: 0.5387 - val_accuracy: 0.7944 - val_auc: 0.6590
Epoch 5/10
53/53 - 0s - loss: 0.4883 - accuracy: 0.8036 - auc: 0.6105
53/53 - 1s - loss: 0.4883 - accuracy: 0.8036 - auc: 0.6105 - val_loss: 0.4909 - val_accuracy: 0.7958 - val_auc: 0.6723
Epoch 6/10
53/53 - 0s - loss: 0.4884 - accuracy: 0.8054 - auc: 0.6146
53/53 - 1s - loss: 0.4884 - accuracy: 0.8054 - auc: 0.6146 - val_loss: 0.4772 - val_accuracy: 0.7944 - val_auc: 0.6681
Epoch 7/10
53/53 - 0s - loss: 0.4736 - accuracy: 0.8030 - auc: 0.6464
53/53 - 1s - loss: 0.4736 - accuracy: 0.8030 - auc: 0.6464 - val_loss: 0.4872 - val_accuracy: 0.7944 - val_auc: 0.6662
Epoch 8/10
53/53 - 0s - loss: 0.4745 - accuracy: 0.8036 - auc: 0.6442
53/53 - 1s - loss: 0.4745 - accuracy: 0.8036 - auc: 0.6442 - val_loss: 0.4766 - val_accuracy: 0.7944 - val_auc: 0.6741
Epoch 9/10
53/53 - 0s - loss: 0.4813 - accuracy: 0.8071 - auc: 0.6217
53/53 - 1s - loss: 0.4813 - accuracy: 0.8071 - auc: 0.6217 - val_loss: 0.4868 - val_accuracy: 0.7986 - val_auc: 0.6860
Epoch 10/10
53/53 - 0s - loss: 0.4705 - accuracy: 0.7970 - auc: 0.6644
53/53 - 1s - loss: 0.4705 - accuracy: 0.7970 - auc: 0.6644 - val_loss: 0.4756 - val_accuracy: 0.7944 - val_auc: 0.6791

#save(fit_train, file="../output/fit_train.RData")

Step 5: Run test on test images

tm_test = NA

feature_test <- array(unlist(dat_test[, -6007]), dim = c(600, 152, 3))

if(run.test){
  #(file="../output/fit_train.RData")
  tm_test <- system.time(
                         {label_pred <- test(fit_train, feature_test, type = "classes");
                          prob_pred <- test(fit_train, feature_test, type = "proba")}
                         )
  
}

label_test <- as.integer(dat_test$label)
weight_test <- rep(NA, length(label_test))
for (v in unique(label_test)){
  weight_test[label_test == v] = 0.5 * length(label_test) / length(label_test[label_test == v])
}

label_test <- ifelse(label_test == 2, 1, 0)
accu <- sum(weight_test * (label_pred == label_test)) / sum(weight_test)
# prob_pred <- apply(prob_pred, 1, max)
prob_pred <- prob_pred[, 1]
tpr.fpr <- WeightedROC(prob_pred, label_test, weight_test)
auc <- WeightedAUC(tpr.fpr)

cat("The accuracy of model:", "CNN", "is", accu*100, "%.\n")

The accuracy of model: CNN is 50.84034 %.

cat("The AUC of model:", "CNN", "is", auc, ".\n")

The AUC of model: CNN is 0.687084 .

Summarize Running Time

Prediction performance matters, so does the running times for constructing features and for training the model, especially when the computation resource is limited.

cat("Time for constructing training features=", tm_feature_train[1], "s \n")

Time for constructing training features= 0.51 s 

cat("Time for constructing testing features=", tm_feature_test[1], "s \n")

Time for constructing testing features= 0.14 s 

cat("Time for training model=", tm_train[1], "s \n") 

Time for training model= 12.73 s 

cat("Time for testing model=", tm_test[1], "s \n")

Time for testing model= 0.14 s 

###Reference - Du, S., Tao, Y., & Martinez, A. M. (2014). Compound facial expressions of emotion. Proceedings of the National Academy of Sciences, 111(15), E1454-E1462.